Ex Parte Takahashi

Patent Trial and Appeal Board

Feb 16, 2017

12989600 (P.T.A.B. Feb. 16, 2017)

United States Patent and Trademark Office
UNITED STATES DEPARTMENT OF COMMERCE
United States Patent and Trademark Office
Address: COMMISSIONER FOR PATENTS
P.O.Box 1450
Alexandria, Virginia 22313-1450
www.uspto.gov
APPLICATION NO. FILING DATE FIRST NAMED INVENTOR ATTORNEY DOCKET NO. CONFIRMATION NO.
12/989,600 10/25/2010 Nobuaki Takahashi Q121450 1199
65565 7590
SUGHRUE-265550
2100 PENNSYLVANIA AVE. NW
WASHINGTON, DC 20037-3213
EXAMINER
HUYNH, PHUONG N
ART UNIT PAPER NUMBER
1644
NOTIFICATION DATE DELIVERY MODE
02/21/2017 ELECTRONIC
Please find below and/or attached an Office communication concerning this application or proceeding.
The time period for reply, if any, is set in the attached communication.
Notice of the Office communication was sent electronically on above-indicated "Notification Date" to the
following e-mail address(es):
SUGHRUE265550@SUGHRUE.COM
PPROCESSING@SUGHRUE.COM
USPTO@sughrue.com
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
BEFORE THE PATENT TRIAL AND APPEAL BOARD
Ex parte NOBUAKI TAKAHASHI
Appeal 2016-002188
Application 12/989,600
Technology Center 1600
Before ERIC B. GRIMES, JEFFREY N. FREDMAN, and
JOHN E. SCHNEIDER, Administrative Patent Judges.
FREDMAN, Administrative Patent Judge.
DECISION ON APPEAL
This is an appeal1 under 35U.S.C. § 134 involving claims to a
multivalent antibody. The Examiner rejected the claims as failing to comply
with the written description requirement, as not enabling the full scope of the
claims, and as obvious. We have jurisdiction under 35 U.S.C. § 6(b). We
affirm the obviousness rejection but reverse the written description and
nonenablement rejections.
1 Appellant identifies the Real Party in Interest as KYOWA HAKKO KIRIN
CO., LTD (see App. Br. 2).
Appeal 2016-002188
Application 12/989,600
Statement of the Case
Background
“Immunoglobulins are glycoproteins which exist in serum, and tissue
and body fluid of all mammals and have a function of recognizing foreign
antigens” (Spec. 12). “The basic structure of immunoglobulin is composed
of two L (light) chains and two H (heavy) chains. Class and subclass of the
immunoglobulin are determined by H chains. Different class and subclass
of immunoglobulins are known to have different functions” (Id.).
[A] multivalent antibody which recognizes different antigens
was produced by a hybrid hybridoma. However, when this
method is used, since two different types of heavy chain and
light chain are expressed in one cell, approximately ten
combinations of the heavy chain and the light chain of an
antibody are obtained. Accordingly, as a result, the
productivity of a multivalent antibody having desired
combination of a heavy chain and a light chain is lowered, and
moreover it is difficult to isolate and purify the targeted
multivalent antibody.
(Spec. 14).
“[T]he present inventors have found that, among multivalent
antibodies comprising multiple antigen recognition sites in one heavy chain
polypeptide, a multivalent antibody comprising antigen recognition sites
which are not close to each other is highly stable and excellent in
productivity” (Spec. 110).
2
Appeal 2016-002188
Application 12/989,600
The Claims
Claims 1, 5—8, 13, and 14 are on appeal. Claim 1 is representative and
reads as follows:
1. A multivalent antibody comprising four antigen
recognition sites, wherein said multivalent antibody comprises
four light chain polypeptides each comprising a light chain
variable region, and wherein said multivalent antibody further
comprises two heavy chain polypeptides each comprising two
heavy chain variable regions linked to each other via a linker,
wherein each of said heavy chain variable regions is
complexed with one of said light chain polypeptides to form the
antigen recognition sites, wherein said multivalent antibody
binds to two different epitopes,
and wherein said linker consists of a CHI domain
originating from an IgM or IgG4 subclass, or consists of a CHI
domain fragment of 14 or more amino acids in which position
14 of said fragment is a cysteine, said fragment originating
from an IgM or IgG4 subclass.
The issues
A. The Examiner rejected claims 1, 5, 13, and 14 under 35 U.S.C.
§ 103(a) as obvious over Miller,2 Beckmann,3 and Moore4 (Ans. 20-25).
B. The Examiner rejected claims 1, 5—8, 13, and 14 under 35 U.S.C.
§112, first paragraph, as failing to comply with the written description
requirement (Ans. 2—11).
C. The Examiner rejected claims 1, 5—8, 13, and 14 under 35 U.S.C.
§112, first paragraph, enablement (Ans. 11—19).
2 Miller et al., US 2006/0025576 Al, published Feb. 2, 2006 (“Miller”).
3 Beckmann, R., WO 2007/085814 Al, published Aug. 2, 2007
(“Beckmann”).
4 Moore et al., WO 2005/087810 A2, published Sept. 22, 2005 (“Moore”).
3
Appeal 2016-002188
Application 12/989,600
A. 35 U.S.C. § 103(a) over Miller, Beckmann, and Moore
The Examiner finds Miller teaches a “multivalent antibody that
comprises four light chains represent[ed] by VLCL and two heavy chains
represented] by VHCH1-VHCH1-CH2-CH3” with the “two heavy chain
variable region (VH) linked to each other via a linker represented] by XI”
(Ans. 20). The Examiner finds that XI may “consist of a CHI domain”
from an IgG (Ans. 21).
The Examiner acknowledges that Miller does not teach IgG4 CHI
domains, but finds that Beckmann and Moore both teach the use of IgG4
CHI domains in fusion proteins (see Ans. 21—22). The Examiner finds the
use of IgG4 CHI domains in Miller obvious
because the substitution of one known element (CHI domain
from IgGl isotype) for another (i.e., CHI domain from human
IgG4 or IgM isotype) would have yielded predictable results
(i.e., capable of forming dimer by non-covalently pairing with
an immunoglobulin light chain .kappa, or ,lam[b]da. constant
region) to one of ordinary skill in the art at the time of the
invention.
(Ans. 24).
The issues with respect to this rejection are:
(i) Does the evidence of record support the Examiner’s conclusion
that Miller, Beckmann, and Moore render claim 1 obvious?
(ii) If so, has Appellant presented evidence of secondary
considerations, that when weighed with the evidence of obviousness, is
sufficient to support a conclusion of non-obviousness?
4
Appeal 2016-002188
Application 12/989,600
Findings of Fact
1. Miller teaches “multivalent antibodies (e.g. tetravalent
antibodies) with three or more antigen binding sites, which can be readily
produced by recombinant expression of nucleic acid encoding the
polypeptide chains of the antibody” (Miller 1 60).
2. Miller teaches “multispecific antibodies, i.e., antibodies that
have binding specificities for at least two different epitopes or antigenic
determinants” (Miller 1213).
3. Figure 4B of Miller is reproduced below:
Heavy*chain
Figure 4B depicts “schematically tetravalent antibodies according to the
present invention . . . the dimeriza[ti]on domain of the tetravalent antibody is
an Fc region” (Miller 175).
5
Appeal 2016-002188
Application 12/989,600
4. Miller teaches:
Where the multivalent antibody comprises an Fc region,
preferably, the three or more antigen binding sites are provided
amino terminal to the Fc region .... This may be achieved by
providing a first polypeptide chain represented by the formula
VD1-Xl-VD2-X2-Fc, wherein (1) VD1 is a first heavy or light
chain variable domain (preferably a heavy chain variable
domain), (2) VD2 is a second heavy or light chain variable
domain (preferably a heavy chain variable domain), (3) Fc
comprises one chain of an Fc region, and (4) XI and X2
represent an optional intervening amino acid or polypeptide.
Preferably XI and X2 comprise, or consist of, a CHI domain
(Miller 1197).
5. Miller teaches, in Example 1, “a tetravalent anti-HER2
antibody, called an ‘Octopus antibody’ (OctHER2)” (Miller 1393) that
comprised CHI linkers between VH regions (see Miller, Fig. 5) where the
data showed the antibody contained four binding sites (see Miller 1398).
6. Beckmann teaches “recombinant fusion proteins that contain
natural junctions. The fusion proteins of the invention comprise at least two
portions derived from two different polypeptides, and at least one natural
junction between the two portions” (Beckmann 2:19—22).
7. Beckmann teaches “a recombinant fusion protein comprising a
natural junction . . . whereby the likelihood of inducing an immune response
is reduced in comparison to a corresponding fusion protein that does not
contain a natural junction” (Beckmann 16:7—9).
8. Beckmann teaches “[pjreferably, the fusion protein of this
embodiment comprises a human antibody constant domain, such as an IgG
CHI domain” (Beckmann 5:30—31).
6
Appeal 2016-002188
Application 12/989,600
9. Beckmann teaches “[preferably, the antibody heavy chain
constant domain is a human antibody heavy chain constant domain. In
particular embodiments, the carboxy-terminus of the hybrid antibody
variable domain is bonded directly to IgG CHI or IgG CH2 (e.g., IgGl CHI,
IgG4 CHI, IgGl CH2, IgG4 CH2)” (Beckmann 41:15-18).
10. Moore teaches “[ijmmunoglobulin CHI domains include, for
example, the CHI domains of human . . . gamma4 (SEQ ID NO:51)”
(Moore 6:19—21).
11. Moore teaches the “immunoglobulin CHI domain is capable of
non-covalently pairing with an immunoglobulin light chain k or X constant
region. As disclosed above, the CHI domain can be a CHI domain from the
y, a, p, s, or 8 classes of immunoglobulins, or a variant of a wild-type
domain. Within the y class, a CHI domain from any of the yl, y2, y3, or y4
subclasses can be used” (Moore 14:28—32).
Principles of Law
“The combination of familiar elements according to known methods
is likely to be obvious when it does no more than yield predictable results.”
KSRInt’l Co. v. Teleflex Inc., 550 U.S. 398, 416 (2007).
Analysis
We adopt the Examiner’s findings of fact and reasoning regarding the
scope and content of the prior art (Ans. 2—5; FF 1—11) and agree that the
claims are rendered obvious by Miller, Beckmann, and Moore. We address
Appellant’s arguments below.
7
Appeal 2016-002188
Application 12/989,600
Prima Facie case
Appellant contends that “none of the ’576 [Miller], ’814 [Beckmann]
and ’810 [Moore] Publications even remotely suggest to persons of ordinary
skill in the art to use a CHI domain sequence from IgG4 or IgM as the linker
sequence to link two heavy chain variable regions” (App. Br. 24).
We do not find this argument persuasive. Miller teaches every
element of the claim (FF 1—3) including the use of an Ig CHI domain
linking heavy chain regions (FF 4), and solely lacks a teaching that the Ig
CHI domain should be selected from an IgG4 subclass. Both Beckmann
and Moore teach IgG4 subclass CHI domains as predictable and known
equivalent CHI domains to the other Ig CHI domains of Miller (FF 8—11).
“If a person of ordinary skill can implement a predictable variation, § 103
likely bars its patentability.” KSR, 550 U.S. at 417. Moreover, an “[e]xpress
suggestion to substitute one equivalent for another need not be present to
render such substitution obvious.” In re Fout, 675 F.2d 297, 301 (CCPA
1982).
Appellant contends that
the amino acid sequence “KTHT” included in the “CHI”
portion depicted in Figure 5 of [Miller] actually represents a
partial amino acid sequence of a hinge region. In other words,
the linker connecting the two VH domains in these Octopus
antibodies is not “a linker [which] consists of a CHI domain
originating from an IgM or IgG4 subclass, or consists of a CHI
domain fragment of 14 or more amino acids in which position
14 of said fragment is a cysteine, said fragment originating
from an IgM or IgG4 subclass,” but is composed of the CHI
domain of IgGl, a hinge fragment, and G-S or G-S-G-S.
(App. Br. 25).
8
Appeal 2016-002188
Application 12/989,600
We do not find this argument persuasive because it relies upon a
single example in Miller, not the actual disclosure of Miller, which teaches
“a first polypeptide chain represented by the formula VD1-Xl-VD2-X2-Fc
. . . . Preferably XI and X2 comprise, or consist of, a CHI domain” (FF 4;
emphasis added). With VD1 and VD2 representing heavy and/or light chain
regions (FF 4), Miller clearly suggests XI and X2 linkers that “consist” of a
CHI domain as required claim 1. Disclosed examples and preferred
embodiments do not constitute a teaching away from a broader disclosure or
non-preferred embodiments. In re Susi, 440 F.2d 442, 446 n.3 (CCPA
1971).
Moreover, Appellant’s claim recites an “antibody comprising” and
“polypeptides each comprising” later followed by “linker consists.” Crish
found that the “reasonable interpretation of the claims containing both of the
terms ‘comprising’ and ‘consists’ is that the term ‘consists’ limits the ‘said
portion’ language to the subsequently recited numbered nucleotides, but the
earlier term ‘comprising’ means that the claim can include that portion plus
other nucleotides.” In re Crish, 393 F.3d 1253, 1257 (Fed. Cir. 2004).
Applying Crish to instant claim 1, we interpret the “said linker consists”
limitation to require a CHI linker, but the earlier open transition
“comprising” means that the claim can include additional components
between the two heavy chain variable regions. With this claim
interpretation, even the example of Miller discussed above would, when
substituted with the known IgG4 CHI domains of Beckmann or Moore,
suffice to render the claim 1 obvious.
9
Appeal 2016-002188
Application 12/989,600
Appellant separately argues that “[njothing in [Beckmann] reasonably
suggests the utility of a CHI domain (or fragment thereof) from IgG4 or
IgM as a linker for directly linking a plurality of VH domains together”
(App. Br. 26 (emphasis omitted)) and that Moore “would not have led
persons of ordinary skill in the art to use a linker consisting of a CHI
domain, or fragment thereof, originating from IgM or IgG4, as a linker to
connect two VH domains in a multivalent antibody” (App. Br. 27).
We are not persuaded because Beckmann and Moore are not relied
upon to suggest the use of a CHI linker, an element already expressly taught
by Miller (FF 4), but rather to demonstrate that different CHI types were
known, and in particular, that IgG4 CHI subtypes were known equivalents
(FF 9-11). “Non-obviousness cannot be established by attacking references
individually where the rejection is based upon the teachings of a
combination of references.” In re Merck & Co., 800 F.2d 1091, 1097 (Fed.
Cir. 1986). A reference “must be read, not in isolation, but for what it fairly
teaches in combination with the prior art as a whole.” Id-
Secondary Considerations
Appellant contends
the specification as filed (at, inter alia, Tables 8 and 14) sets
forth unexpected properties for the claimed antibodies, namely
reduced aggregation vis-a-vis when the equivalent CHI domain
or fragment thereof from other antibody isotypes, e.g., IgGl, is
used. This unexpected property is of clear practical
significance, aiding in the development and production of
therapeutic multivalent antibodies.
(App. Br. 30; cf. App. Br. 28—29).
10
Appeal 2016-002188
Application 12/989,600
We do not find this evidence persuasive of unexpected results for
several reasons. First, the comparison is not with the closest prior art of
Miller, who provides specific working examples suitable for comparison (FF
5), but with which no comparisons were made. See In re Baxter Travenol
Labs., 952 F.2d 388, 392 (Fed. Cir. 1991) (“[W]hen unexpected results are
used as evidence of nonobviousness, the results must be shown to be
unexpected compared with the closest prior art.”). Thus, we also find
unpersuasive Appellant’s argument that “it is manifestly improper for the
Office to require a comparison (of the properties of the claimed invention)
against what could be produced” (Reply Br. 12) because Miller’s examples
were actually produced and represent the closest prior art.
Second, the Specification teaches “281 VL4-CH1-281 comprises the
sequence comprising the CHI sequence (positions 118 to 215 of the EU
index of Rabat) of IgG4 and the BamRl sequence (GGATCC) added to its
3’-terminal were inserted as a linker between the heavy chains (3’-terminal
of 281VL4 heavy chain, 5’-terminal of 281 heavy chain)” (Spec. 52:4—7).
This indicates that the experiments were performed with a linker that did not
“consist” of CHI sequence, but also included the additional BamHI
sequence which Appellant has argued is excluded from claim 1 by the
“consisting of’ language. Thus, following Appellant’s reasoning, the
comparison did not use antibodies within the scope of the claims.
Third, the results shown in Tables 8 and 14 (see Spec. 55 and 63—65)
appear to be limited to the use of the CD28 antigen recognition site 281
VL4, but claim 1 encompasses antibodies with any antigen recognition site
whatsoever binding to any epitope. Unexpected results must be
11
Appeal 2016-002188
Application 12/989,600
“commensurate in scope with the degree of protection sought by the claimed
subject matter.” In re Harris, 409 F.3d 1339, 1344 (Fed. Cir. 2005). Here,
Appellant provides no persuasive evidence establishing that the aggregation
results would be expected with the full scope of claim 1, or indeed, with any
other antibody sequences.
Fourth, Appellant does not provide any evidence that those skilled in
the art would recognize the results as unexpected or surprising or other
evidence that the result was unexpected. Because the Specification does not
identify the results as unexpected, Appellant must provide evidence
explaining why the results are, in fact, unexpected. Currently, Appellant
only has attorney argument supporting this point. See In re Soni, 54 F.3d
746, 750 (Fed. Cir. 1995) (“It is well settled that unexpected results must be
established by factual evidence. Mere argument or conclusory statements . .
. [do] not suffice.” (citation omitted)). Also see In re Pearson, 494 F.2d
1399, 1405 (CCPA 1974) (“Attorney’s argument in a brief cannot take the
place of evidence.”).
Conclusion of Law
(i) The evidence of record supports the Examiner’s conclusion that
Miller, Beckmann, and Moore render claim 1 obvious.
(ii) Appellant has not presented evidence of secondary considerations,
that when weighed with the evidence of obviousness, is sufficient to support
a conclusion of non-obviousness.
12
Appeal 2016-002188
Application 12/989,600
B. 35 U.S. C. § 112, first paragraph — Written Description
The Examiner finds:
Since the disclosure does not describe the common attributes or
structural characteristics that identify members of the genus of
antigen recognitions sites, i.e., the six CDRs from heavy and
light chain variable domains that bind to two different epitopes
that encompassed linear or three-dimensional structure, and
because the binding specificity of genus is highly variable, one
of skill in the art would not immediately envisage the genus of
multivalent antibody comprising four antigen recognition sites
that specifically bind to “two different epitopes” ....
(Ans. 5). The Examiner also finds “there is a high degree of unpredictability
in proper peptide linkages between the two variable domains in both heavy
chain and light chain because each DVD-Ig protein is unique and its
properties are often related to or limited by the properties of the parental
mAbs” (Ans. 9).
Appellant contends:
At the time of the invention, a vast array of antibodies had been
characterized (i.e., their variable region sequences and their
binding specificity were known) — and persons of skill in the art
would have understood, without a needless re-description in the
specification, which antibody variable regions could be used to
impart a given antigen-binding specificity (and would have
recognized that they could be employed in tandem in
multivalent antibodies). Thus, because the preexisting
knowledge in this art was substantial, persons of skill in the art
at the time of the invention would have recognized, from the
description in the specification, that Appellant was in
possession of the genus of multivalent antibodies which bind to
“two different epitopes.”
(App. Br. 8).
13
Appeal 2016-002188
Application 12/989,600
The issue with respect to this rejection is: Does the evidence of record
support the Examiner’s conclusion that the Specification fails to provide
descriptive support for the claims?
Findings of Fact
12. Miller teaches:
In order to generate the multivalent antibody, a “parent” or
“starting” antibody with variable domains directed against an
antigen may be prepared using-various methodologies for-
making antibodies .... The sequences of the variable domains
of the starting or parent antibody may be used in the design of
the multivalent antibody....
(Miller 1221).
13. Miller teaches “[pjolyclonal antibodies are preferably raised in
animals by multiple subcutaneous (sc) or intraperitoneal (ip) injections of
the relevant antigen and an adjuvant” and “monoclonal antibodies may be
made using the hybridoma method first described by Kohler” (Miller H 223,
227).
14. Miller teaches a “useful method for identification of certain
residues or regions of the antibody that are preferred locations for
mutagenesis [that] is called ‘alanine scanning mutagenesis’” (Miller 1254).
15. The Specification teaches “the multivalent antibody can be
prepared by, for example, cloning genes of multiple monoclonal antibodies
against different epitopes which are expressed from the hybridoma obtained
as described above, determining antigen recognition sites and designing a
gene of the multivalent antibody containing the obtained antigen recognition
sites” (Spec. 19:9-13).
14
Appeal 2016-002188
Application 12/989,600
16. The Specification teaches the “antigen recognition site can be
obtained by specifying and separating the appropriate antigen recognition
site using a technique such as a phage display method, a yeast display or the
like as well as the above-mentioned method using the hybridoma” (Spec.
22:10-12).
Principles of Law
[T]he determination of what is needed to support generic claims
to biological subject matter depends on a variety of factors,
such as the existing knowledge in the particular field, the extent
and content of the prior art, the maturity of the science or
technology, the predictability of the aspect at issue, and other
considerations appropriate to the subject matter.
Capon v. Eshhar, 418 F.3d 1349, 1359 (Fed. Cir. 2005).
Analysis
We find Appellant has the better position. As in Capon, the instant
invention is not in the particular amino acid sequences that encode the
antigen recognition cites for the multivalent antibody, or even in the amino
acid sequences of the CHI domains, but rather “in the novel combination of
the [amino acid] segments to achieve a novel result.” Capon, 418 F.3d at
1358.
The evidence of record, including the nearly anticipatory Miller
reference cited by the Examiner, demonstrates that multivalent antibodies in
general and antigen recognition sites and epitopes in particular, are very well
known (FF 12—14).
The instant facts are therefore also similar to those in Falko-Gunter,
which teaches that “where, as in this case, accessible literature sources
clearly provided, as of the relevant date, genes and their nucleotide
15
Appeal 2016-002188
Application 12/989,600
sequences (here ‘essential genes’), satisfaction of the written description
requirement does not require either the recitation or incorporation by
reference (where permitted) of such genes and sequences.” Falko-Gunter
Falknerv. Inglis, 448 F.3d 1357, 1368 (Fed. Cir. 2006) (footnote omitted).
We recognize, but find unpersuasive, the Examiner’s finding that the
“specification does not identify [i)] a complete structure, ii) partial structure,
iii) physical and/or chemical properties or iv) functional characteristics”
(Ans. 7). In this case, the prior art and Specification provide the complete or
partial sequences of a very large number of antibodies, whose physical,
chemical, and functional characteristics are both very well-known and easily
determined, if unknown (FF 12—16).
The Examiner’s focus on “binding specificity” is misplaced, because
the claims do not require any particular binding type or degree of binding
specificity, and indeed, simply require the ability to bind to two different
epitopes. Given that Miller teaches multivalent antibodies that do so (FF 2),
we do not find this concern persuasive.
Moreover, we recognize the Colman,5 Dufher,6 Wu,7 and Morrison8
references suggest that there is some degree of unpredictability in antigen
5 Colman, P., Effects of amino acid sequence changes on antibody-antigen
interactions, 145 Research in Immunology 33—36 (1994).
6 Dufher et al., Harnessing Phage and Ribosome Display for Antibody
Optimisation, 24 Trends Biotechnol. 523—529 (2006).
7 Wu et al., Humanization of a Murine Monoclonal Antibody by
Simultaneous Optimization of Framework and CDR Residues, 294
J. Mol. Biol. 151-162 (1999).
8 Morrison S., Two Heads are better than one, 25 Nature Biotechnology
1233-1234 (2007).
16
Appeal 2016-002188
Application 12/989,600
binding and functional activity of antibodies, but these references do not
overcome the clear teachings in the prior art that screening for desired
antibody sequences is routine, and may be identified using a variety of well-
known art recognized approaches (FF 12—16).
Conclusion of Law
The evidence of record does not support the Examiner’s conclusion
that the Specification fails to provide descriptive support for the claims.
C. 35 U.S. C. § 112, first paragraph — Enablement
The Examiner finds:
Given the lack of guidance as the binding specificity and
the universe of multivalent antibodies that bind to any two
different epitopes from infinite number of antigens, the lack of
specific guidance as to the structure, i.e., immunoglobulin
heavy and light chain comprising the 6 CDRs shared by all
antibodies, the insufficient working example and the
unpredictable art, it would take undue experimentation to come
up with the combination of four light chain and two heavy
chains bind to which two different epitopes that encompassed
3-D structure and either activates (agonist) or inactivate
(antagonist) upon binding to which antigen, in turn, effective
for diagnostic or therapeutic use, i.e., treating various diseases
including cancer.
(Ans. 16).
Appellant contends “the examined claims are product claims - which
do not recite or require that the antibody is used for treating a disease
through activation/inactivation of the antigen to which it binds. . . . [T]he
claimed antibodies have other utilities, including in vitro utilities” (App. Br.
13).
17
Appeal 2016-002188
Application 12/989,600
We find that Appellant has the better position. As we apply the
Wands factor analysis, the only specific functional requirement for the
multivalent antibody of claim 1 is the ability to bind two different epitopes.
Claim 1 is broadly drawn to any multivalent antibodies with four
antigen recognition sites that contain the specific linkers, such as the IgG4
CHI domain linker. However, the Specification provides guidance on
generating the multivalent antibodies (FF 15—16). The prior art of Miller,
Wu, and Morrison teach and exemplify the generation of multivalent
antibodies (FF 5; Wu 159, col. 2; Morrison 1233, fig. 1). The Specification
has working examples (see, e.g., Spec. 70-71, Example 37, “Preparation of
Multivalent Antibody Expression Vector comprising Antigen Recognition
Site for CD28 (281 VL4 Origin) at N-terminal Side, Antigen Recognition
Site for CD40 (281 Origin) at C-terminal Side and Linker”). There is no
dispute that the skill in the art is high.
Wands itself dealt with the generation of hybridomas for generation of
monoclonal antibodies. In re Wands, 858 F.2d 731, 738 (Fed. Cir. 1988).
The Court found that in spite of the low success rate for any individual
experimental screening result, “in the monoclonal antibody art it appears that
an ‘experiment’ is not simply the screening of a single hybridoma, but is
rather the entire attempt to make a monoclonal antibody against a particular
antigen.” Id. at 740.
The same reasoning applies here. While some experimentation may
be required to combine particular known prior art antibody sequences to
form multivalent antibodies, including some screening using phage display
or other processes, the Examiner has not established that such screening
18
Appeal 2016-002188
Application 12/989,600
would have been undue. In particular, the evidence is not persuasive of a
substantial degree of unpredictability in obtaining a multivalent antibody
capable of binding two epitopes based on the teachings of the prior art and
Specification. We therefore reverse the nonenablement rejection.
SUMMARY
In summary, we affirm the rejection of claim 1 under 35 U.S.C.
§ 103(a) as obvious over Miller, Beckmann, and Moore. Claims 5,13, and
14 fall with claim 1.
We reverse the rejection of claims 1, 5—8, 13, and 14 under 35 U.S.C.
§112, first paragraph as failing to comply with the written description
requirement.
We reverse the rejection of claims 1, 5—8, 13, and 14 under 35 U.S.C.
§112, first paragraph, enablement.
No time period for taking any subsequent action in connection with
this appeal may be extended under 37 C.F.R. § 1.136(a).
AFFIRMED-IN-PART
19